Post-infection Irritable Bowel Syndrome

INTRODUCTION

With a worldwide prevalence ranging between 7–21%, irritable bowel syndrome (IBS) is one of the most commonly diagnosed gastrointestinal (GI) conditions.¹ It is characterized by chronic abdominal pain associated with bowel disturbances such as diarrhea, constipation, or both. Although the etiology of IBS remains to be fully clarified, the heterogeneous pathophysiology involves peripheral (gut) and central (spinal, supra spinal) factors operating in a bidirectional manner. Growing literature has provided a deeper understanding of the alterations involving the mucosal barrier, immune function, gut microbiota, enteroendocrine and visceral sensorimotor function.² Additionally, genetics and environment (diet, infections, and psychosocial factors) appear to play a role in the predisposition and manifestations of IBS.^{3, 4}

Acute infectious gastroenteritis (bacterial, viral and protozoal) has been shown to be one of the strongest risk factors for development of IBS⁵, a condition referred to as post-infection IBS (PI-IBS). Although described in literature on medical diseases of the war in 1917⁶, the first formal description of PI-IBS was made in 1962 by Chaudhary and Truelove.⁷ Every year, 1 in 6 adults suffer from an episode of foodborne illness in the United States, placing a great number of individuals at risk of developing PI-IBS.⁸ Conservative estimates through mathematical modeling suggest that PI-IBS prevalence in the community could be 9%, accounting for over half the overall prevalence of IBS in the United States.⁹

DIAGNOSIS

Recently, the Rome Foundation Working Group proposed diagnostic criteria for PI-IBS based on Rome IV criteria (Box 1).¹ The symptoms typically develop immediately after resolution of acute infectious gastroenteritis. The gastroenteritis episode can be diagnosed either by a positive stool culture or clinically by the presence of ≥2 of fever, vomiting or diarrhea.¹⁰ In order to aid diagnosis and management, PI-IBS is further classified into four subtypes on the basis of the predominant bowel pattern graded on the Bristol stool scale: diarrhea-predominant IBS (IBS-D), constipation-predominant (IBS-C), mixed bowel habits (IBS-M) and unclassified (IBS-U). A bowel pattern is considered predominant when it is present ≥¼ of the time as either hard/lumpy (IBS-C), loose/watery (IBS-D) or both (IBS-M). In a recent meta-analysis of studies from 1994–2015 including over 21,000 patients that had an episode of infectious enteritis, IBS-M and IBS-D were found to be the most common phenotypes, accounting for 46% and 40% of the cases, respectively.⁵

Box 1. Diagnostic Criteria for Post-Infection Irritable Bowel Syndrome Based on Rome IV.

1. Recurrent abdominal pain, on average, at least 1 day per week in the last 3 months, with symptom onset at least 6 months before diagnosis, associated with 2 or more of the following*:

   1. Related to defecation
   2. A change in frequency of stool
   3. A change in form (appearance) of stool

Supportive criteria

1. Infectious gastroenteritis defined by either:

   1. A positive culture in a symptomatic individual
   2. Presence of at least 2 of the following acute symptoms: fever, vomiting and diarrhea

2. Should not meet criteria for IBS before onset of acute illness

3. Symptom development either immediately after resolution of acute infectious gastroenteritis or within 30 days of resolution of acute symptoms

* Criteria fulfilled for the last 3 months with symptom onset at least 6 months before diagnosis.

The burden of chronic GI symptoms post-infection may even be greater than what is captured by Rome criteria. In a recent study, we demonstrated that in addition to 21% prevalence of PI-IBS following Campylobacter enteritis, an additional 9% suffered from new-onset abdominal pain, bowel disturbances or both, without meeting the Rome criteria for IBS.¹¹ In addition to the development of new-onset IBS, a GI infection can also result in changes in pre-existing IBS phenotype. Approximately 50% of IBS-C switched to either IBS-M or IBS-D after an episode of Campylobacter enteritis in our study.¹¹ It is unclear if the infection is causative in this switch as IBS subtypes have been known to switch over time.¹² Furthermore, 38% of the patients showed an increase in pain frequency post-infection.¹¹ Overall, these data suggest that GI infections can result in new onset of chronic GI symptoms as well as changes in pre-existing chronic GI symptoms.

DIFFERENTIAL DIAGNOSIS

The diagnosis of PI-IBS should be primarily based on the clinical features, reserving diagnostic tests when alarm features for an organic disease are present.¹ A positive diagnostic approach, rather than by exclusion, has been demonstrated to lower health care costs while increasing patient satisfaction.¹³ Limited diagnostic testing including complete blood count, C-reactive protein, celiac serology and fecal calprotectin may be considered based on clinical suspicion.¹⁴ Small intestinal bacterial overgrowth (SIBO) is a condition where an excessive growth of microorganisms occurs in the small intestine. Patients present with a variety of symptoms, including bloating, flatulence, abdominal discomfort and diarrhea. It can be diagnosed with a jejunal aspirate demonstrating >1×10³ colony forming units per ml or with carbohydrate breath testing. Patients with IBS have 3 times higher odds for SIBO than healthy individuals as demonstrated in two meta-analysis.^{15, 16} However, the prevalence of SIBO in PI-IBS is unknown and routine testing for SIBO in PI-IBS is not recommended. Tropical sprue also shares similarities with PI-IBS and is triggered by a GI infection.¹⁷ In contrast to PI-IBS, enteropathy in tropical sprue predominantly involves the small bowel and results in malabsorption of carbohydrates, fat, folate and vitamin B12 which can be tested using specific assays (D-xylose, fecal fat, and vitamin B12 level), and enteropathy confirmed with histologic findings in a duodenal biopsy.¹⁸ Interestingly, in the last 20 years, reports on tropical sprue have decreased, while they have increased for PI-IBS.¹⁹ Given the similarities, some have hypothesized that in certain geographical settings, these might reflect spectrum of a disease process initiated by an infectious insult.²⁰ Prevalence of Giardia in PI-IBS ranged from 38–80%, although only one of the studies confirmed eradication.^21–23 Symptoms of giardiasis such as diarrhea, abdominal cramps, and flatulence can mimic those of PI-IBS. These are often due to lactose malabsorption.²⁴ Hence, chronic giardiasis should be excluded in patients with suspected Giardia PI-IBS.

Celiac disease can have a clinical picture similar to PI-IBS. In a recent meta-analysis of 36 studies, patients with suspected IBS, regardless of the subtype, had 3–4 times odds of having a positive celiac disease test (serologic or histologic) compared to asymptomatic individuals.¹⁴ Another condition with potential overlap of symptoms with PI-IBS is microscopic colitis. A recent systematic review showed that in individuals meeting criteria for IBS-D, the prevalence of microscopic colitis is 10%.²⁵ As of yet, no specific studies for PI-IBS exist, but given the high prevalence of the diarrhea subtype in PI-IBS, it is an important differential diagnosis to take into consideration.²⁶ However, microscopic colitis usually presents at ages older than 50 years, whereas PI-IBS is higher in the younger. Additionally, microscopic colitis patients have painless chronic watery diarrhea (commonly with a Bristol Stool Score ≥6), whereas PI-IBS is accompanied with abdominal pain and Bristol Stool Score of 4–5. Random colonic biopsies may be warranted to diagnose microscopic colitis when clinical suspicion is high. According to a study that evaluated colonic biopsies in patients with non-IBS-C, the presence of inflammatory bowel disease (IBD) was 0.4%, similar to healthy controls.²⁷ However, acute enterocolitis has been associated with an increased risk of IBD. One study reported a hazard ratio of 1.9 in the first year after a Salmonella or Campylobacter infection, which increased to 2.9 over 7.5 years of follow-up.²⁸ Another study demonstrated an odds ratio of 1.4 for development of IBD after an infectious enteritis (viral, bacterial and protozoal). Risk was 5 times higher in patients with a pre-existing IBS.²⁹ We do not recommend routine work-up for exclusion of IBD unless there are suggestive symptoms such as weight loss, GI bleeding, fever, tenesmus, etc. The diagnostic approach for PI-IBS is outlined in Figure 1.

Figure 1. Diagnostic and therapeutic approach for post-infection irritable bowel syndrome.

The presence of severe symptoms, significant weight loss, abnormal physical findings or testing when diagnosing PI-IBS, as well as lack of treatment response after 4–8 weeks, warrants considering for other etiologies. Further diagnostic tests should be selected based on clinical suspicion. Pharmacologic treatment recommendations include focus on IBS-M and IBS-D which account for the vast majority of PI-IBS patients. Antispasmodics (peppermint oil, dicyclomine, hyoscine, pinaverium), TCAs (amitryptiline, imipramine, desipramine), probiotics (Bifidobacterium sp., Lactobacillus plantarum), antidiarrheals (loperamide, eluxadoline), bile acid sequestrants (cholestyramine, colesevelam, colestipol), 5HT₃ antagonists (ondansetron, alosetron, ramosetron). In the case of IBS-C, the treatment recommendations for the altered bowel habit are soluble fibers (psyllium), osmotic laxatives (polyethylene glycol, lactulose) and intestinal secretagogues (linaclotide, lubiprostone, plecanatide). Abbreviations: FODMAP, fermentable oligosaccharides, disaccharides, monosaccharides and polyols; TCAs, tricyclic antidepressants; 5HT₃, 5-hydroxytryptamine type 3; IBD, inflammatory bowel disease. Modified from Barbara G, Grover M, Bercik P, et al. Rome Foundation working team report on post-infection irritable bowel syndrome. Gastroenterology 2019;156:46–58 e7.

EPIDEMIOLOGY AND NATURAL HISTORY

Viral (Norovirus, Rotavirus)^{30, 31}, bacterial (Campylobacter, Salmonella, Escherichia coli, Shigella, Clostridium difficile)^{11, 32–35} and protozoal (Giardia)^21–23 enterocolitis have been associated with PI-IBS with varying degrees of risk. In our meta-analysis, the overall point prevalence of PI-IBS in studies looking at 12 months and >12 months post-infection was 10% and 15%, respectively (Figure 2).⁵ Exposed individuals were 4.2 times more likely to develop PI-IBS than unexposed individuals within 12 months of infection. Risk decreased beyond 12 months, however, it remained significantly higher compared to individuals without history of infection.⁵ There was no significant difference in prevalence comparing Rome I, II and III.⁵ None of the studies included have used Rome IV thus far, which may result in lower prevalence as noted with general IBS.³⁶ It is hypothesized that due to the high incidence of infectious gastroenteritis and poor recall of milder episodes, the true prevalence of PI-IBS may be underestimated.¹⁰ While comparing different pathogens, protozoal enteritis shows the highest risk for PI-IBS development, followed by bacterial, and, lastly, viral. The significant decrease in PI-IBS prevalence from 19% to 4% one year after viral infection may be due to the less invasive nature of the pathogen, perhaps avoiding a stronger host response.⁵ Pathogen-based risk over time is summarized in Figure 2.

Figure 2. Natural history of post-infection irritable bowel syndrome.

Within 1 year after an acute gastrointestinal infection (bacterial, viral, parasitic and overall), there is a 4-fold increase in risk for development of irritable bowel syndrome compared to non-exposed individuals. Risk decreases beyond 1 year of infection, but remains significantly increased, except in viral enteritis. *Non-significant. Abbreviations: RR, relative risk (compared to unexposed derived from same population). Not all studies had unexposed controls; hence prevalence and RR estimates not from completely overlapping studies. Data from Klem F, Wadhwa A, Prokop LJ, et al. Prevalence, risk factors, and outcomes of irritable bowel syndrome after infectious enteritis: A systematic review and meta-analysis. Gastroenterology 2017;152:1042–54 e1.

Demographic, psychological and clinical factors related to the acute infection have been associated with PI-IBS risk. Female and younger patients are more likely to develop PI-IBS.⁵ Higher anxiety, depression, somatization and neuroticism scores preceding or during acute enteritis increase the risk, possibly by altering host or pathogen responses.⁵ Clinical characteristics of the acute enterocolitis such as longer duration of diarrhea, presence of bloody stools, abdominal cramps and hospitalization significantly increase the risk.^{4, 5, 11} Fever was not observed to be a risk factor.⁵ In fact, our study showed that it might be protective¹¹, possibly reflecting a protective host response to the injury. Antibiotic use has been shown to be a risk factor, however, most of those studies are from outside the U.S. where the overall use of antibiotics was considerably lower (5–14%)^{37, 38}, perhaps representing a bias between health-care seeking and subsequent PI-IBS. In other studies, including those from the United States where antibiotic use is higher (47%−77%), no association with PI-IBS risk was seen.^{11, 39} Understanding and modeling these clinical risk-factors may allow identification of high-risk cohorts that are most suitable for strategies to prevent PI-IBS development.

Functional dyspepsia (FD) is another common functional GI disorder that causes epigastric pain, postprandial fullness and early satiety. Its development after infectious gastroenteritis has been reported⁴⁰, with a prevalence of ~9% and a RR of 2.5 in a systematic review of 19 studies.⁴¹ An overlap between PI-IBS and PI-FD can exist as demonstrated post-giardiasis where 26% developed PI-FD, 85% of who also met criteria for PI-IBS.⁴²

PATHOPHISIOLOGY

In the following sections, we will highlight changes in the gut microbiome, epithelium and neuronal excitation that have been either demonstrated or postulated to play a role in PI-IBS pathophysiology (Figure 3).

Figure 3. Alterations in the intestinal environment underlying the pathophysiology of post-infection irritable bowel syndrome.

The development of PI-IBS after acute infectious enterocolitis may be due to number of changes in intestinal microbiome, mucosal, immune or neuronal function. Microbiota changes include decreased diversity and increased Firmicutes:Bacteroides ratio. These may change luminal milieu by altering composition of bile acids, bile salts and proteases. Epithelial changes such as increased density of enteroendocrine cells, increased serotonin availability may increase intestinal motility. An altered mucosal barrier function can contribute towards immune dysregulation and neuronal hypersensitivity. Furthermore, immunophenotypic changes such as increased mast cell density, increased Th1/Th2 cell ratio and expression of proinflammatory cytokines can mediate chronic gut dysfunction as well as neuronal excitability. Lastly, inflammasome mediated depletion of neurons and nervous remodeling can affect motility and secretion. Abbreviations: 5-HT, 5-hydroxytryptamine; EC, enterochromaffin; HTRE3E, 5-hydroxytryptamine receptor 3E; IL, interleukin; PAR-2, protease activated receptor-2; SERT, serotonin transporter; TJ, tight junction; VGLUT2, vesicular glutamate transporter 2; ZO-1, zonula occludens-1; Th1/Th2, T helper type 1/2; TNF-α, tumor necrosis factor alpha.

Gut microbiome

A healthy microbiome, although difficult to characterize, is an interconnected community that serves a wide array of functions in the gut ranging from digestion and production of valuable nutrients, shaping host immunity, xenobiotic metabolism and protection from pathogens and exogenous antigens. The microbiome is resilient to external stressors such as GI infection, antibiotic exposure, and dietary changes.⁴³ Firmicutes, Bacteriodetes and Actinobacteria are the dominant microbial taxa; however, in infectious enteritis and subsequently in PI-IBS, significant disruption to the core microbiome has been demonstrated.⁴⁴ PI-IBS patients demonstrated a 12- fold increase in Bacteroidetes and decrease in Firmicutes, and Clostridiales compared to healthy individuals.⁴⁵ Travelers’ diarrhea patients also demonstrated a decrease in the Firmicutes: Bacteriodetes ratio and changes in β-diversity compared to healthy controls.⁴⁶ Similarly, studies from both Britain and Sweden found a significant increase in Bacteroides and decrease in Clostridiales.^{47, 48} Moreover, an individual’s susceptibility to developing acute infection as well as PI-IBS may also depend on microbiota composition prior to infection. In abattoir workers, increased abundance of Bacteroides in pre-employment stool samples predicted a greater risk of C. jejuni infection during employment.⁴⁹ Further work needs to be done to understand how the microbiome prior to infection may shift after acute gastroenteritis and if it can serve as biomarker or predictor for PI-IBS development.

An example of how microbiota changes may influence host physiology is by altering the intestinal protease milieu. Proteases are important mediators of intestinal barrier dysfunction and visceral hypersensitivity, both robustly associated with IBS.^{50, 51} Higher fecal proteolytic activity (PA) associates with greater symptom severity and lower microbial diversity in patients with C. jejuni PI-IBS.⁵² Moreover, microbiota changes can also mediate changes in bile salts as well as cause bile acid malabsorption in PI-IBS patients which can induce diarrhea.⁵³ These observations highlight some plausible mechanisms by which changes in intestinal microbiome can mediate the pathophysiology of PI-IBS.

One of the most commonly used models is inducing Trichinella spiralis infection in rodents. Many of the pathophysiological mechanisms associated with PI-IBS in humans, like immune dysfunction^54–56, hypersensitivity⁵⁷, and changes in colonic motility⁵⁸, have been demonstrated in this model. Lactobacillus and Bifidobacterium administration caused a reduction in visceral sensitivity, an increase in tight junction (TJ) protein expression and suppression of pro-inflammatory cytokines (IL-6 and IL-17) when compared to control treated animals.^{54, 55, 59} Using the same model, daily fecal microbial transplantation (FMT) over the course of a week resulted in improved barrier function and a higher pain threshold compared to controls.⁶⁰ Interestingly, administration of Bifidobacterim longum alone was just as effective as FMT suggesting individual taxa may be sufficient for restoring gut function.⁶⁰ Similar decreases in pro-inflammatory cytokine production were seen after the addition of Lactobacillus in both ileum and colonic ex vivo organ cultures from PI-IBS human tissue.⁶¹ Additionally, FMT in humans proved effective in treating PI-IBS caused by Giardia for 7-weeks; however the effect did not last as symptoms recurred at 1 year follow-up.⁶² Further studies investigating the short- and long-term efficacy of microbial replenishment in PI-IBS are needed to determine the specific taxa or consortium of taxa that can induce an effect as well as plausible mechanisms.

Epithelial changes

Acute gastroenteritis has immediate effects on the intestinal epithelium such as inflammation, edema, and possibly hemorrhaging, with PI-IBS causing pathogens known to illicit varying degrees of epithelial injury.¹⁰ Long-term epithelial changes have been reported in PI-IBS patients. Small non-coding, miRNA-510 was found to be down-regulated in mucosal biopsies of PI-IBS patients which correlated with increased pro-inflammatory cytokines.⁶³ Gene expression changes have been seen in the serotonin reuptake transporter (SERT), a critical transporter for the metabolism of serotonin (5-HT), with PI-IBS patients having significantly lower expression which will increase luminal availability of 5HT⁶⁴ where stimulation of 5-HT receptors on nerve endings by ligands triggers neural stimuli. Experimental PI-IBS rats administered L. rhamnosus supernatant had an increase in colonic SERT mRNA and protein expression indicating microbial modulation can restore serotonergic imbalance.⁶⁵ Enterochromaffin (EC) cells release 5-HT and have chemo and mechanosensory roles. In both Shigella and Campylobacter PI-IBS, there is an increase in density of colonic EC cells (up to 25%).^66–68 In contrast, patients with post Giardia IBS were found to have lower EC cell counts compared to controls, but increased cholecystokinin (CCK) immunoreactive enteroendocrine cells.⁶⁹ Patients with Shigella PI-IBS also have increased ileal and rectal Il-1β expression, as well as ileal mast cell infiltration.⁷⁰ In cases of PI-IBS attributable to viral infection, the effect on the GI epithelium remains less clear.

Mucosal permeability:

Four human studies assessing intestinal permeability found an evidence of greater permeability in PI-IBS.^{67, 71–73} One study showed 5 of 10 Campylobacter PI-IBS patients had increased permeability;⁶⁷ however, a much larger study from the Walkerton outbreak reported an increase in permeability in only 16% of the PI-IBS patients studied.⁷² Epithelial barrier integrity is dependent on cell-to-cell adhesion which is regulated by the presence of tight junctions. Genetic variations in E-cadherin (CDH1) associated with PI-IBS risk in the Walkerton cohort.³ In rodent models, barrier disruption and bacterial translocation correlated with occludin and claudin-4 degradation, along with decreased zona occludens and increased myosin light chain kinase (MLCK) expression, a mediator of barrier dysfunction.^{74, 75} Moreover, PI-IBS patients with high fecal proteolytic activity (PA) have shown increased in vivo and ex vivo distal gut permeability which depended on the level of fecal PA.⁵²

Immune responses:

Acute infectious gastroenteritis results in the recruitment of immune cells to the gut. Cytokines IL-1β, IL-10, IL-13 and INFγ are increased in PI-IBS patients compared to controls long after acute infection.^{70, 76, 77} Compared to healthy controls, colonic biopsies from PI-IBS patients release pro-inflammatory cytokines when exposed to commensal bacteria ex vivo.⁷⁶ This observation suggests colonic mucosa from PI-IBS may inappropriately respond to commensal bacteria, thereby creating a milieu of prolonged immune dysregulation. In fact, mast cells, macrophages and monocytes^{34, 67} along with T lymphocytes, and intraepithelial lymphocytes (IEL) have all been found to be elevated in PI-IBS up to 5 years post-infection^{66, 67}. Colonic supernatants from PI-IBS patients can activate mast cells which are an established mediator of visceral hypersensitivity in IBS.^{78, 79} Unspecified infections resulting in PI-IBS have reported an increase in EC and T cells, but no changes in IEL cells.⁶⁶ Significantly increased proportions of CD4+ T-cells and double positive CD4/CD8+ cells have been reported in the lamina propria of mucosal biopsies from PI-IBS patients.⁸⁰ CD3+ and CD8+ lymphocytes, IELs and CD68+ macrophages are increased in Shigella PI-IBS patients compared to controls as well.⁶⁸

PI-IBS patients appear to have a shift in the T-helper response with an increased ratio of Th1/Th2 with increased IFNγ and decreased IL-10 levels, respectively.⁷⁷ In a perpetual state of disequilibrium, elevated Th1 cells may mediate pain through effects on intestinal permeability. In a PI-IBS mouse model, inhibition of MLCK caused significant decreases in the production of Th1 cytokines, decreased visceral hypersensitivity and restoration of intestinal barrier function suggesting a mechanistic link between barrier and sensory dysfunction.⁷⁵ In contrast, some studies have found no differences in mast cell or IEL counts in PI-IBS cases compared with control biopsies.^{64, 79} This indicates that more needs to be done to understand immune dysfunction in PI-IBS beyond simplistic assessment of cell counts.

Neuronal activation

Commensal microbes as well as enteric pathogens have been shown to influence the excitability of enteric–associated neurons (EAN), neurons critical for monitoring and maintaining homeostasis within the gut by facilitating nutrient absorption and colonic motility⁸¹, as well as the activation of immune cells.⁸² A recent study demonstrated that through NLRP6 inflammasome and caspase 11 activation, acute infections in mice caused a 20–30% reduction in intrinsic ileo-colonic EANs 7 days post infection, with neuronal loss persisting up to 126 days post infection. This was associated with delayed intestinal transit time.⁸³ This loss was only seen for the excitatory vesicular glutamate transporter (VGLUT2⁺) neurons and not for the neuronal nitric oxide synthase (nNOS) or somatostatin positive inhibitory neurons. Interestingly, restoration of a healthy microbiome in these mice resulted in the recovery of EANs; however, it was not reported whether transit time normalized. Additionally, B. longum has recently been shown to reduce visceral hypersensitivity in T. spiralis infected mice by specifically inhibiting the NLRP3 inflammasome.⁸⁴ Extensive nerve remodeling was reported in Nippostrongylus brasiliensis infected rats⁸⁵, accompanied by recruitment and activation of mast cells.⁸⁶ Activated mast cells surrounding nerve fibers in the ileum have been reported in PI-IBS patients⁷⁰ and released tryptase can mediate neuronal excitation through cleavage of protease activated receptor-2 (PAR-2).⁸⁷ Activation of PAR-2 is known to increase permeability and visceral sensitivity.⁵¹ In a mouse model of PI-IBS, visceral hypersensitivity was reduced after the administration of a PAR-2 antagonist.⁸⁸ This is further supported by the observation that excitability of dorsal root ganglion neurons is absent in PAR-2 knockout mice after incubation with colonic supernatants from IBS-D patients.⁸⁹ A pro-nociceptive change was reported in the rectal biopsies of PI-IBS patients as a result of sensitization of the ion channel, transient receptor potential vanilloid 1 (TRPV1) sensitization. Submucosal neuronal activation was significantly increased in PI-IBS patients compared to healthy controls, even two years after initial infection was cleared.⁹⁰ Recently, EphrinB2/ephB2 receptor tyrosine kinase, an enzyme important for regulating neuronal activation and excitation, was found to be up-regulated in the colonic muscularis of T. spiralis infected rats. Colonic hypersensitivity and hypercontractility were ameliorated upon administration of ephB2Fc, an ephB2 blocker, indicating a critical role for ephrinB2/ephB2 signaling in facilitating neuronal maturation and potentiation in the gut.⁹¹

MANAGEMENT

Currently, treatment options for PI-IBS are limited with no specific FDA-approved agents. Therefore, current management strategies of PI-IBS are based on expert opinion. As with IBS in general, treatment should be guided based on phenotype and predominant symptoms.¹⁰

Patient Education and Reassurance

Effective doctor-patient relationship and communication are essential in the management of PI-IBS. The first step involves educating the patient about the link between the infection and the development of chronic GI symptoms. Patients should be reassured that PI-IBS tends to have a more benign course and symptoms tend to improve and sometimes disappear over time in many patients, especially when the etiology is viral.^{5, 92}

Dietary modifications

Given that a large proportion of patients with PI-IBS have IBS-D or IBS-M, a reasonable initial approach is a 4–8-week trial of low fermentable oligosaccharides, disaccharides, monosaccharides and polyols (FODMAPs), as it has been demonstrated to improve symptoms in IBS-D.⁹³ A gluten-free diet in IBS patients without celiac disease has not been found to be effective. However, one randomized controlled trial showed that reintroduction of gluten lead to rebound of symptoms in IBS patients previously symptomatically controlled on a gluten-free diet.⁹⁴ Given that wheat contains fructans, which are oligosaccharides, a trial comparing gluten-free with a low FODMAPs diet alone found no difference in response, suggesting that the benefit from the gluten-free diet is likely due to the lower ingestion of oligosaccharides.⁹⁵

Pharmacologic agents

A summary of recommended pharmacologic treatments is included in Figure 1. One of the few treatment strategies that has been tested specifically for PI-IBS is glutamine. A randomized controlled trial studying PI-IBS-D patients with increased in vivo intestinal permeability found 80% of patients reporting a ≥50-point reduction in the IBS symptom severity score, as compared to only 6% in the placebo group. Other significant findings were an improvement of bowel movement frequency and consistency, as well as a normalization of intestinal permeability.⁹⁶ If replicated, this could serve as a key therapeutic option for PI-IBS patients. A randomized, placebo-controlled trial of mesalazine 4 g/day showed no benefit in symptom relief of patients with IBS-D. However, in a small subgroup that met criteria for PI-IBS (n=13), a significant improvement in abdominal pain, urgency and stool consistency was observed.⁹⁷ In another double-blind controlled trial including 17 patients with diarrhea predominant PI-IBS, mesalamine failed to show improvement of symptoms and quality of life.⁹⁸ However, in an uncontrolled study evaluating 389 patients affected in a large outbreak of hemorrhagic enterocolitis by Shiga-like toxin-producing E. coli in Germany, the use of mesalazine aimed to reduce intestinal inflammation during the acute infection appeared to significantly protect against PI-IBS development.⁹⁹ A randomized controlled trial of prednisone focused primarily on evaluating changes in rectal biopsies in PI-IBS, observed a significant reduction of T- lymphocyte counts, however, no changes were observed in EC count and overall symptom improvement.¹⁰⁰

SUMMARY

Intestinal infections can significantly increase the risk of IBS, a chronic and morbid GI disorder with high health-care utilization. Epidemiological and mechanistic studies in humans as well as animal models point towards specific host-pathogen interactions that may lead to the development of this chronic sequel post-infection. These advancements in understanding will be helpful in triaging individuals who are at high risk as well as designing targeted pharmacotherapy. Microbial restoration, augmentation of barrier function and targeting visceral hypersensitivity remain the most promising areas for therapeutic intervention. From a clinical standpoint, recognition and communication of this condition by physicians will result in better patient outcomes and lower costs. Much needs to be done in expanding the implications of the findings seen in human biopsies and models of PI-IBS, as well as designing proof-of-concept clinical trials specific to the PI-IBS population. The information gained from this will help target this chronic complication of intestinal infections and will also have broader implications for understanding and treating IBS, one of the commonest GI diagnoses.

Synopsis:

Growing epidemiological data supports that acute gastrointestinal infection is one of the strongest risk factors for development of irritable bowel syndrome (IBS). The risk of post-infection IBS (PI-IBS) appears to be more with bacterial and protozoal than viral enterocolitis. Younger individuals, females and those with severe enterocolitis are more likely to develop PI-IBS. Disease mechanisms in animal models and humans involve chronic perturbation of intestinal microbiome, epithelial and neuronal remodeling, and immune activation. These can lead to luminal (increased proteolytic activity, altered bile acid composition) and physiological (increased permeability, transit changes, and visceral hypersensitivity) alterations which can mediate PI-IBS symptoms.

CLINICS CARE POINTS.

• Acute infectious gastroenteritis leads to development of IBS in approximately 1 in 10 individuals. Risk is highest with parasitic infections, followed by bacterial, and lastly viral.

• Main risk factors are younger age, female sex, psychological distress and a greater severity of infection. Identifying patients at risk is of importance to give appropriate follow-up and treatment.

• Follow a positive diagnostic approach rather than reaching diagnosis by exclusion. Reserve limited testing (complete blood count, C-reactive protein, celiac serology and fecal calprotectin) for cases where clinical suspicion of another diagnosis is high.

• Patient education regarding a GI infection acting as a trigger of IBS, as well as reassurance that PI-IBS symptoms tend to resolve in a subset of patients is the first step in management.

• Although some treatment strategies have been tested specifically in PI-IBS, current management recommendations and extrapolated from routine IBS.

• Glutamine may be an effective therapy for PI-IBS patients; however, currently can only be used in off-label settings.

• Lack of response to initial treatment after 4–8 weeks, worsening or severe symptoms may warrant the need to rule out other causes.

Key Points:

• Acute infectious gastroenteritis is a strong risk-factor for development of irritable bowel syndrome (IBS). Approximately, 90% of post-infection IBS (PI-IBS) is either mixed or diarrhea predominant.

• Younger age, female sex, infection severity and psychological distress are associated with greater risk of PI-IBS. Risk also varies with pathogen type; protozoal having the highest prevalence, followed by bacterial and lastly viral.

• Key pathophysiological mechanisms in PI-IBS include microbial dysbiosis, mucosal barrier dysfunction, immune dysregulation and neuronal hypersensitivity.

• Although some treatment strategies like mesalamine, corticosteroids and glutamine have been tested specifically for PI-IBS, current management recommendations for PI-IBS are extrapolated from routine IBS.